kano (OP)
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
March 06, 2017, 07:30:26 AM |
|
I just run Avalon 1.5 days. The invaid of Canaan too high. Just for the new block, the invalid if 2 A741 and 1 A721 in a chain is 1%. Invalid of R4 is 0%. How to fix the invalid and how invalid affect your luck pool or my hashing and sharing?
The invalid is stale share? Do I get paid for that? And how to avoid that?
... no you don't get paid for stale shares ... Just like we don't get paid for a stale block coz it's worthless. Meanwhile, I spent some time going though the bmminer code to see if they were still faking their stale stats by "hiding" the stale shares. The reason being to compare your stales since if one miner is reporting them correctly and the other one is faking them ... well then it really doesn't mean much comparing them. However, your hash rate shown on the web site of course is only based on valid, non-stale shares, so the hash rate is of course what matters. Anyway, The bitmain driver code used to "attempt" to intercept stale shares and discard them rather than pass them back to the main cgminer code. i.e. it was hiding them so they weren't counted. The main cgminer code of course should decide based on the cgminer options, not the bitmain driver. That was one of the fixes I did to the S1 driver years ago when I first fixed the bitmain crappy code, that bitmain never copied any of the fixes. Anyway, while perusing the bmminer driver code I felt the urge to run away screaming ... Here's their latest function that screws around with the shares before passing them back to the main cgminer code ... https://github.com/bitmaintech/bmminer/blob/master/driver-btm-c5.c#L7887Damn, who let those retard programmers at bitmain near a keyboard. What it is supposed to do is simply retrun the nonce and let the main cgminer code check it properly, not all that screwing around they are doing there. But that wasn't the reason for wanting to run away screaming ... There is a function in all your recent bitmain miners "send_mac" https://github.com/bitmaintech/bmminer/blob/master/driver-btm-c5.c#L7764It connects to bitmain https://github.com/bitmaintech/bmminer/blob/master/driver-btm-c5.c#L7683Since AUTH_URL is: https://github.com/bitmaintech/bmminer/blob/master/driver-btm-c5.c#L223"auth.minerlink.com" and it sends your mac address, some miner id, and some hash board id string to bitmain https://github.com/bitmaintech/bmminer/blob/master/driver-btm-c5.c#L7774Randomly every 1 to 11 minutes while the miner is running https://github.com/bitmaintech/bmminer/blob/master/driver-btm-c5.c#L7783It doesn't matter what pool you are mining on, it still connects to bitmain Now here's the real fun part ... It checks for a reply from bitmain and if bitmain replies with anything containing the word "false" https://github.com/bitmaintech/bmminer/blob/master/driver-btm-c5.c#L7755Then it will abort the checking function https://github.com/bitmaintech/bmminer/blob/master/driver-btm-c5.c#L7777with the message "Stop mining!!!" But if it fails to connect to bitmain or it doesn't get a "false" reply, it will continue checking randomly every 1 to 11 minutes connecting to bitmain. Now I wonder why they would have a function like that, there in the code ... ... ... ... ...
|
|
|
|
matscol
Newbie
Offline
Activity: 49
Merit: 0
|
|
March 06, 2017, 08:06:19 AM |
|
Would it make sense to try to fork the code and update the miner with a new version that eliminates the send_mac and setup_send_mac_socket functions entirely?
|
|
|
|
kano (OP)
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
March 06, 2017, 08:25:51 AM |
|
Would it make sense to try to fork the code and update the miner with a new version that eliminates the send_mac and setup_send_mac_socket functions entirely? I got booted out of the cgminer team - so don't ask me Ask the cgminer guy who bitmain pays for pretty much all the code he writes
|
|
|
|
kano (OP)
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
March 06, 2017, 09:35:48 AM |
|
Block! by temptrol A7v1 1.76BTC txn fees
|
|
|
|
Lanet
Newbie
Offline
Activity: 3
Merit: 0
|
|
March 06, 2017, 09:43:10 AM Last edit: March 06, 2017, 10:11:00 AM by Lanet |
|
Kano recommend to buy now? So I have many problems with the S9 antminer. I read that the T9 were more efficient than the s9 in relation to temperature. Comparative temperature S9 vs T9 https://www.youtube.com/watch?v=yHJzo7A312c
|
|
|
|
kano (OP)
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
March 06, 2017, 09:57:49 AM Last edit: March 06, 2017, 10:13:50 AM by kano |
|
Kano recommend to buy now? So I have many problems with the s9 antminer. I read that the t9 were more efficient than the s9 in relation to temperature.
Well I have certain bias to Canaan Avalon as is obvious, also coz they are more reliable according to the few people I have spoken to with a lot of each. I've no idea how the price/power costs are any more since they've all changed over the past month or so, so you'd have to check and compare, but also take into consideration dealing with returns, and lost hashing time, and each companies record on handling that.
|
|
|
|
ComputerGenie
|
|
March 06, 2017, 10:23:42 AM |
|
... I read that the T9 were more efficient than the s9 in relation to temperature. ...
6 chips less per board will always create less heat.
|
If you have to ask "why?", you wouldn`t understand my answer. Always be on the look out, because you never know when you'll be stalked by hit-men that eat nothing but cream cheese....
|
|
|
GWhisper
|
|
March 06, 2017, 10:26:16 AM |
|
I might just block "auth.minerlink.com" at DNS and my router in preparation for when it starts to resolve? Or perhaps it's just something they utilise at hashnest to manage their miners.
|
|
|
|
kano (OP)
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
March 06, 2017, 10:38:12 AM |
|
Block! by BlockBadger S9v2 (prolly a T9) 2.4BTC txn fees
|
|
|
|
GWhisper
|
|
March 06, 2017, 10:45:41 AM |
|
Block! by BlockBadger S9v2 (prolly a T9) 2.4BTC txn fees Good stuff - let's keep'em coming.
|
|
|
|
Newko
|
|
March 06, 2017, 11:46:58 AM |
|
@Citronic: Welcome back!
The Avalons 721s x 120; are mining here at Kano. The S9s v1 x 108; are mining somewhere else. Both farms are powered by hydroelectricity at Cryptoboreas Inc., Labrador-NFL, Canada. Yep, I know just the Avalons are here but I wasn't sure how long they've been here so I just wanted to welcome you back. Maybe you've been back for a while and I missed that (my bad, if that's the case) but good to have you here with lovely Juvia. Spring is just around the corner (in the Northern Hemispere) so I guess Winter Juvia will be bidding us farewell soon. Mine on.
|
|
|
|
kano (OP)
Legendary
Offline
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
|
|
March 06, 2017, 12:43:39 PM |
|
I might just block "auth.minerlink.com" at DNS and my router in preparation for when it starts to resolve? Or perhaps it's just something they utilise at hashnest to manage their miners. Well if it's used by hashnest then they are morons (as usual) The cgminer API has a broadcast option that I wrote in it ... many years ago. You enable it in cgminer --api-mcast Then send out a broadcast message on your network, and all running cgminers with it enabled will answer.
|
|
|
|
clgrissom3
Legendary
Offline
Activity: 1722
Merit: 1032
Carl, aka Sonny :)
|
|
March 06, 2017, 02:35:53 PM |
|
|
|
|
|
miner0007
Jr. Member
Offline
Activity: 74
Merit: 1
|
|
March 06, 2017, 02:45:45 PM |
|
See... our hashrate went down and we are getting more lucky and getting more blocks. we need to keep our hash rate below 27,000TH.
|
Lisk - Javascript Blockchain Development Platform Get more info below if you want to earn MONEY with Lisk https://liskelite.com
|
|
|
NomadGroup
|
|
March 06, 2017, 03:44:03 PM |
|
Oh yeah! The last one was a MONSTER!!!!!!
|
|
|
|
huntingthesnark
|
|
March 06, 2017, 04:04:43 PM |
|
See... our hashrate went down and we are getting more lucky and getting more blocks. we need to keep our hash rate below 27,000TH. Yeah, crazy innit. Hashrate up 30%, no luck at all. Dips a bit and boom. Almost makes you want to mistrust the maths!
|
|
|
|
Biodom
Legendary
Offline
Activity: 3920
Merit: 4400
|
|
March 06, 2017, 04:15:29 PM |
|
See... our hashrate went down and we are getting more lucky and getting more blocks. we need to keep our hash rate below 27,000TH. A logical fallacy of this type: Post hoc ergo propter hocMy dog wakes up and then, almost immediately, sun rises as well. Wow! My dog has the ability to make sun to rise.
|
|
|
|
citronick
Legendary
Offline
Activity: 1834
Merit: 1080
---- winter*juvia -----
|
|
March 06, 2017, 04:24:59 PM |
|
@Citronic: Welcome back!
The Avalons 721s x 120; are mining here at Kano. The S9s v1 x 108; are mining somewhere else. Both farms are powered by hydroelectricity at Cryptoboreas Inc., Labrador-NFL, Canada. Yep, I know just the Avalons are here but I wasn't sure how long they've been here so I just wanted to welcome you back. Maybe you've been back for a while and I missed that (my bad, if that's the case) but good to have you here with lovely Juvia. Spring is just around the corner (in the Northern Hemispere) so I guess Winter Juvia will be bidding us farewell soon. Mine on. Mining since January 2017 those A7s. They are not Juvias though but Grays; Juvia's main squeeze Yes, Summer Gray avatar should be coming soon. Great to see the pool back to her old self.
|
If I provided you good and useful info or just a smile to your day, consider sending me merit points to further validate this Bitcointalk account ~ useful for future account recovery...
|
|
|
miner0007
Jr. Member
Offline
Activity: 74
Merit: 1
|
|
March 06, 2017, 04:30:56 PM Last edit: March 06, 2017, 04:44:03 PM by miner0007 |
|
See... our hashrate went down and we are getting more lucky and getting more blocks. we need to keep our hash rate below 27,000TH. A logical fallacy of this type: Post hoc ergo propter hocMy dog wakes up and then, almost immediately, sun rises as well. Wow! My dog has the ability to make sun to rise. Amazing Dog you have there, but don't be too happy, since my dog has the same capability to make the sun rise. But that is slightly different. I was obviously joking 1st of all, but actually perhaps not!!!! It could be some bug or some threshold or limit which does actually effect it. Perhaps with more hashrate the servers are getting overwhelmed and not sending out the info fast enough, I don't know, I am talking out of my ass, but if it is consistently like that, then something is wrong, not saying that is is consistently like that... But.... Basically saying, don't jump to conclusions that it is indeed just bad luck and nothing more than that!!!
|
Lisk - Javascript Blockchain Development Platform Get more info below if you want to earn MONEY with Lisk https://liskelite.com
|
|
|
Biodom
Legendary
Offline
Activity: 3920
Merit: 4400
|
|
March 06, 2017, 04:44:10 PM |
|
A have a question to kano or some other blockchain or db specialist.
I was under impression that once tx is on the blockchain, even an unconfirmed one, then it is unchangeable. is this really so, or does current soft allow for some sort of "replacement in kind" with the same numerical outputs for currently unconfirmed transactions?
|
|
|
|
|